Tripo AI 3D Generator
Mini toolsOrnith 1.0 9B
Ornith-1.0-9B is the compact, edge-deployable member of the Ornith-1.0 family, spanning 9B to 397B parameters, specialized for agentic software engineering. Post-trained on Qwen 3.5 using a self-improving reinforcement learning framework that co-optimizes both solution rollouts and the task-specific scaffolds guiding them, enabling the model to discover better orchestration strategies without human-designed harnesses. As a reasoning model, it produces structured thinking traces before final answers. It natively supports OpenAI-compatible tool calling and function calling, making it compatible with agent frameworks including OpenHands, OpenCode, and llama.cpp. Runs on a single 80GB GPU in bf16 format (~19GB) with a context window up to 262K tokens. Achieves 69.4% on SWE-bench Verified and 43.1% on Terminal-Bench 2.1, matching or exceeding much larger models like Gemma 4-31B. MIT licensed with no regional restrictions.
New
Text
Gen 7
Overview
Open-source 9B-parameter language model specialized for agentic coding tasks. Post-trained on Qwen 3.5 using a self-improving RL framework that jointly learns to generate solutions and task-specific scaffolds. Achieves state-of-the-art results on SWE-bench Verified (69.4%) and Terminal-Bench 2.1 among comparable models. MIT licensed.
About DeepReinforce
DeepReinforce is an AI research startup founded by Dr. Jiwei Li, focused on using reinforcement learning to build agentic AI systems for coding and system optimization. They developed GrandCode (ranked #1 in Codeforces live competitions, beating all human grandmasters), Ornith-1.0 (open-source LLMs for agentic coding, 9B–397B parameters), and IterX (agentic code optimizer surpassing NVIDIA's cuBLAS).
View Company Profile
Last updated: June 30, 2026
